colwise.md
In dplyr: A Grammar of Data Manipulation

colwise utils gives meaningful error messages

Code
  (expect_error(tbl_at_vars(iris, raw(3))))
Output
  <error/rlang_error>
  Error:
  ! `.vars` must be a character/numeric vector or a `vars()` object, not a raw vector.
Code
  (expect_error(tbl_if_vars(iris, list(identity, force), environment())))
Output
  <error/rlang_error>
  Error:
  ! `.predicate` must have length 1, not 2.
Code
  .funs <- as_fun_list(list(identity, force), caller_env())
  (expect_error(tbl_if_vars(iris, .funs, environment())))
Output
  <error/rlang_error>
  Error:
  ! `.predicate` must have length 1, not 2.